Hi,
well if you look at the entire pipeline, it may well be something like this:
yuv->resize to 640x480->RGB->GL texture upload->lanczos filtering to
320->(either on screen, or bus reading for file writing)
You may have 2 resize ops and a GPU upload (and download?) which are all
expansive.
I can
Hello,
I use QTKit to grab an image from my webcam. Then i need to resize it into
a smaller image.
Going from My CIImage, i use two filter "CILanczosScaleTransform" and "CICrop"
to get my image at the correct size. (640x480 to 320x240+crop).
Doing only that, my CPU go up to 30% of usage while it
